Back to PLS HelpSignificance determination
Posted on 03/21/08 02:38:37
Number of posts: 1
Tomas Kasparek MD PhD
Dep. of Psciatry, Masaryk University
Brno, the Czech Republic
Dear all, allow me a newbie question:
I want to determine the significance of LVs, I perform 10000 permutations. I also wanted to know the reliability and performed the bootstrapping.
Then in results I view "Permuted values greater than observed for a particular LV and get LV: 0.033. From what I read about permutations I understand that this LV is significant at the level of p < 0.033. Is it correct?
What puzzles me is that when I view the "Bootstrap ratio", another p value is given: "Approx. P value", which for the same LV is 0.2. What does this value mean?
I was not able to find answer in the User Guide. Could you clarify it to me or point me to relevant literature?
Thank you in advance!
Tomas Kasparek
I want to determine the significance of LVs, I perform 10000 permutations. I also wanted to know the reliability and performed the bootstrapping.
Then in results I view "Permuted values greater than observed for a particular LV and get LV: 0.033. From what I read about permutations I understand that this LV is significant at the level of p < 0.033. Is it correct?
What puzzles me is that when I view the "Bootstrap ratio", another p value is given: "Approx. P value", which for the same LV is 0.2. What does this value mean?
I was not able to find answer in the User Guide. Could you clarify it to me or point me to relevant literature?
Thank you in advance!
Tomas Kasparek
Replies:
Untitled Post
I'm Online
Posted on 03/31/08 10:37:45
Number of posts: 291
quote:
Dear all, allow me a newbie question:
I want to determine the significance of LVs, I perform 10000 permutations. I also wanted to know the reliability and performed the bootstrapping.
Then in results I view "Permuted values greater than observed for a particular LV and get LV: 0.033. From what I read about permutations I understand that this LV is significant at the level of p < 0.033. Is it correct?
What puzzles me is that when I view the "Bootstrap ratio", another p value is given: "Approx. P value", which for the same LV is 0.2. What does this value mean?
I was not able to find answer in the User Guide. Could you clarify it to me or point me to relevant literature?
Thank you in advance!
Tomas Kasparek
First of all, the PLSgui User's Guide only solves you "How to use PLSgui program". To better understand PLS method, please refer to the articles indicated or other references. It may also need some statistics knowledge. You can find this from the PLSgui User's Guide:Dear all, allow me a newbie question:
I want to determine the significance of LVs, I perform 10000 permutations. I also wanted to know the reliability and performed the bootstrapping.
Then in results I view "Permuted values greater than observed for a particular LV and get LV: 0.033. From what I read about permutations I understand that this LV is significant at the level of p < 0.033. Is it correct?
What puzzles me is that when I view the "Bootstrap ratio", another p value is given: "Approx. P value", which for the same LV is 0.2. What does this value mean?
I was not able to find answer in the User Guide. Could you clarify it to me or point me to relevant literature?
Thank you in advance!
Tomas Kasparek
This User's Guide intend to show
how to use PLSgui program to those users who have already understood the PLS
method. If PLS is new to you, please read relevant
articles first.
To answer your first question, it is almost correct. More precisely: The LV of permuted datamat has a significance at the level of p < 0.033.
To answer your second question: The "approx. P Value" under BS Ratio Plot window depends on the threshold you selected. It is the standard error function of the threshold. Therefore, it is not the P Value you mentioned. Unfortunately, I am not able to point you the relevant literature.
Login to reply to this topic.
